- Replacing `BaseData.AdjustResolution` for `DefaultResolution` and
`SupportedResolutions`
- Making `Resolution` nullable for `Algorithm.AddData` methods
- The `ISubscriptionDataConfigService` will set the default resolution
if none was provided and assert it is supported
- Fix bug with `PythonData` `IsSparseData` and `RequiresMapping`
resolution
- Adding `Security.NullLeverage` value to determine when the
`SecurityInitializer` leverage should be used or not
- Adding regression algorithm which reproduces the issue
Per review comments provided in #3743 regarding the
training/long-running scheduled events and the leaky
bucket algorithm. See the PR for more information.
- Adds PyObject overload to `ScheduleManager.TrainingNow` and `ScheduleManager.Training`
- Adds `QCAlgorithm.Train` helper method
- Adds Python algorithm showing how to use the helper method.
- Creates `CustomUniverseSelectionModel` that mimics `QCAlgorithm.AddUniverse(String, Func<DateTime, IEnumerable<string>>)`
- Replaces `BaseETFUniverse` for `InceptionDateUniverseSelectionModel` that inherits from `CustomUniverseSelectionModel`
- ETF Basket USMs inherits from `InceptionDateUniverseSelectionModel`
Updates usages of GetBaseDataInstance that have a config in scope to use this
new method. This method properly assigns the symbol to the newly created base
data instance. This enables some of the new flag methods to work as expected.
Updated other poor usages where we weren't setting the symbol property.
See #3618
* Added better documentation for AddData methods
* Added new regression algorithms for adding in OnSecuritiesChanged
* Changed regression algorithms to add data that exists
* Styling and logging fixes
* Deleted regression algorithms because they tested behavior similar to
other existing regression algorithms
* Fixed new bug in regression algorithm due to AddData changes
* Added unit tests for wrapt version and package existence
* Fix issue where data would be set to raw normalization mode
* Implement method precedence hack on two additional python methods
* Changed behavior of SymbolCache loading with a ticker
- Previously, we would search the cache with ticker for types that did not
require mapping.
* Added new documentation
* Code cleanup
* Change first date to be SID.DefaultDate if SecurityType isn't expected
* Modified AddData tests to meet expectations. AddData unit tests are passing
* Added new method for Symbol to allow creation with underlying
* Added new unit tests
* Alter SecurityIdentifier method signature for BaseData
* AddData changes to accept underlying Symbol
* Added AddDataImpl
This is being done in an effort to prevent symbol collisions within the
custom data (SecurityType.Base) namespace. The custom data type's name,
is used for disambiguation. As written, this change will break several
user algorithms that still rely on using the implicit string -> Symbol
lift. Providing this type information is optional an currently only being
used by AddData<T> methods. Other consumers of SecurityType.Base symbols
arn't at risk for collision, such as the UserDefinedUniverse, ScheduledUniverse
and others that are LEAN controlled. In order to maintain backwards compatibility,
the SymbolCache was updated to do a hard search when the requested ticker was
not found, looking for the prefix ('ticker.').
Fixes#3332
- Custom data types will know whether or not Lean should use map files
- Updating regression test with sample custom data using map files,
which can run locally
- Adding unit tests for the `SubscriptionDataReaderHistoryProvider`,
checking it mappes equities and options correctly
Adds static methods of the form Parse.<TypeName>(string str) that use
CultureInfo.InvariantCulture. These are to be used when parsing strings.
It's still safe (from the CA1304/CA1305 perspective) to use the ToDecimal
extension method for decimals.
Adds string extension methods for common operations that will now require
CultureInfo.InvariantCulture. These are to be used when converting values
to strings, such as ToStringInvariant()/ToStringInvariant(format), but also
useful for searching within strings, StartsWithInvariant, EndsWithInvariant
and IndexOfInvariant.
FxCop has various rulesets for enforcing things within our codebase.
For this particular issue, we'll be enforcing CA1304 and CA1305 to
ensure we're always using an IFormatProvider or a CultureInfo where
applicable.
Linked Issue: #3045
The `NullReferenceException` type is intended to only be thrown by the CLR.
In most cases, it should be converted to an `ArgumentException` or an
`InvalidOperationException`, depending on if the null value is a parameter
to the current method or not.
The `Exception` type should never really be thrown as it doesn't provide any
additional information or hints as to the issue. It also forces users that
would like to handle expected exceptions to catch all exceptions. These are
converted to an exception type that more accurately describes the reason for
raising the exception: `KeyNotFoundException`, `InvalidOperationException`
- Setting SPY as the default security benchmark
- The security benchmark subscription will be added at `UniverseSelection`
as an internal subscription. Using its own dedicated Security instance
which doesn't live in the algorithms.Securities collection.
- Reducing algorithms exposure to internal subscriptions
- `TimeSliceFactory` will prioritize higher resolution bars, when same
symbol is present twice (for non-internal subscriptionst)
- Adding regression test `CustomUniverseWithBenchmarkRegressionAlgorithm`
- We will now check if python selection method returned `Universe.Unchanged`
- Removing `ToList()` call on fine and coarse data before sending it to
the python algorithm
- Adding regression algorithms
In order to provide full Lean Indicator functionality to python custom indicators, they need to inherit from a C# class. `PythonIndicator` will serve for this purpose.
Algorithms can use the former version (no inheritance).
In order to add support custom python indicators for `QCAlgorithm.PlotIndicator`, we created a `PythonIndicator` class that wraps the custom python indicator. In `QCAlgorithm`, the reference of the wrapper is saved into a dictionary keyed by the python indicator handle.
- Adding `IRealTimeHandler.OnSecurityChanged()` will be used to update
the `OnEndOfDay` security related scheduled events
- Adding `BaseRealTimeHandler.cs` to reduce code duplication in the
`Backtesting` and `LiveTrading` `RealTimeHandlers`
- Adding CSharp and Python regression tests
- Deprecating `OnEndOfDay()` callback because of two reasons, mainly
because Python does not support two methods with the same name, but also
because different assets have different market close times.
- `ScheduledEvents` set at the same time will now be deterministic
- Adding `LazyToUpper()` implementation, that will avoid the call to
`ToUpper` if the string is already upper.
- Reduce the timezone conversions at `Time.EachTradeableDayInTimeZone`
- `TotalPortfolioValue` will iterate over all securities once
- Adding new `SecurityIdentifier` cache, significant impact for
algorithms using coarse/fine data
- `TimeSliceFactory` will avoid creating empty collections
- `ExecutionModels` will check target collection count before trying to
enumerate
- Reduce calls to .`TotalPortfolioValue`
- `SecurityValues` will only be created when required
- `TimeKeeper` will use TimeZone unique Id as dictionary key. The
TimeZone hash is expensive.
- `AlgorithmManager` will avoid calling `DateTime.UtcNow`,
`ConvertFromUtc()` and `RoundDownInTimeZone()`
Martin Molinero